采用空间和通道激励注意力机制优化ResNet-50的CFRP/TC4叠层材料钻削刀具磨损状态监测

doi:10.3969/j.issn.1004-132X.2024.10.009

摘要/Abstract

摘要： 针对碳纤维增强复合材料(CFRP)与钛合金组成的叠层材料在制备装配孔时存在刀具磨损严重的问题，提出了一种空间和通道激励注意力机制(scSE)优化深度残差神经网络(ResNet-50)的刀具磨损监测方法。开展钻削实验，采集钻削过程中的力和温度信号，信号经连续小波变换转换为小波尺度谱。搭建ResNet-50网络结构，从空间和通道双维度对卷积提取的特征图进行权重标定。研究结果表明，scSE可以从空间和通道两个维度做到增强有用特征，抑制无用特征，经scSE优化的网络结构识别准确度达到96.15%。

关键词: 刀具磨损, 连续小波变换, 空间和通道激励注意力机制, 深度残差神经网络

Abstract: Aiming at the problems of severe tool wear in the preparation of assembly holes for laminated materials consisting of carbon fibre reinforced composites and titanium alloys, a tool wear monitoring method with scSE optimised deep residual neural network(ResNet-50) was proposed. Drilling experiments were carried out to collect force and temperature signals during the drilling processes. The signals were converted to wavelet scale spectrum using continuous wavelet transform. The ResNet-50 network structure was built to calibrate the weights of the convolutionally extracted feature maps from both spatial and channel dimensions. The results show that scSE may enhance the useful features and suppress the useless features from both spatial and channel dimensions, and the recognition accuracy of the network structure optimised by scSE reaches 96.15%.

Key words: tool wear, continuous wavelet transform, spatial and channel excitation attention mechanism(scSE), deep residual neural network

中图分类号:

TH164
V257

聂鹏1, 杨程越1, 彭新月1, 于家鹤2, 潘五九1. 采用空间和通道激励注意力机制优化ResNet-50的CFRP/TC4叠层材料钻削刀具磨损状态监测[J]. 中国机械工程, 2024, 35(10): 1793-1801.

NIE Peng1, YANG Chengyue1, PENG Xinyue1, YU Jiahe2, PAN Wujiu1. Tool Wear Condition Monitoring for Drilling CFRP/TC4 Laminated Materials Using scSE Optimised ResNet-50[J]. China Mechanical Engineering, 2024, 35(10): 1793-1801.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.cmemo.org.cn/CN/10.3969/j.issn.1004-132X.2024.10.009

http://www.cmemo.org.cn/CN/Y2024/V35/I10/1793

参考文献

［1］白大山, 陈五一, 陈雪梅. 碳纤维增强复合材料/轻合金叠层结构制孔技术研究进展［J］. 航空制造技术, 2022, 65(9):82-90.
BAI Dashan, CHEN Wuyi, CHEN Xuemei. Progressof Hole Making Technology for Carbon Fibre Reinforced Composites/Light Alloy Laminated Structures［J］. Aeronautical Manufacturing Technology, 2022, 65(9):82-90.
［2］刘献礼, 李雪冰, 丁明娜, 等. 面向智能制造的刀具全生命周期智能管控技术［J］. 机械工程学报, 2021, 57(10):196-219.
LIU Xianli, LI Xuebing, DING Mingna, et al. Intelligent Control Technology for the Whole Life Cycle of Cutting Tools Towards Intelligent Manufacturing［J］. Journal of Mechanical Engineering, 2021, 57(10):196-219.
［3］ZHANG X W, YU T B, XU P F, et al. In-process Stochastic Tool Wear Identification and Its Application to the Improved Cutting Force Modeling of Micro Milling［J］. Mechanical Systems and Signal Processing, 2022, 164:1-13.
［4］孔凡霞, 张德远. 高温合金微量润滑振动钻削温度与刀具磨损［J］. 北京航空航天大学学报, 2012, 38(6):849-852.
KONG Fanxia, ZHANG Deyuan. High Temperature Alloy Micro-lubrication Vibratory Drilling Temperature and Tool Wear［J］. Journal of Beijing University of Aeronautics and Astronautics, 2012, 38(6):849-852.
［5］祝和意, 李宏波, 张合沛, 等. 基于改进CRITIC法的TBM刀具声发射信号研究［J］. 振动与冲击, 2016, 35(6):197-202.
ZHU Heyi, LI Hongbo, ZHANG Hepei, et al. Study of Acoustic Emission Signals from TBM Tools Based on The Improved CRITIC Method［J］. Journal of Vibration and Shock, 2016, 35(6):197-202.
［6］滕洪钊, 邓朝晖, 吕黎曙, 等. 多传感器信息融合的加工过程状态监测研究［J］. 机械工程学报, 2022, 58(6):26-41.
TENG Hongzhao, DENG Chaohui, LYU Lishu, et al. Research on Multi-sensor Information Fusion for Process Condition Monitoring［J］. Journal of Mechanical Engineering, 2022, 58(6):26-41.
［7］刘会永, 张松, 李剑峰, 等. 采用改进CNN-BiLSTM模型的刀具磨损状态监测［J］.中国机械工程, 2022, 33(16):1940-1947.
LIU Huiyong, ZHANG Song, LI Jianfeng, et al. Tool Wear Detection Based on Improved CNN-BiLSTM Model［J］. China Mechanical Engineering, 2022, 33(16):1940-1947.
［8］GOUARIR A, MARTNEZ G, TERRAZAS G, et al. In-process Tool Wear Prediction System Based on Machine Learning Techniques and Force Analysis［J］. Procedia CIRP, 2018, 77:501-504.
［9］QIN B, WANG Y Q, LIU K, et al. A Novel Online Tool Condition Monitoring Method for Milling Titanium Alloy with Consideration of Tool Wear Law［J］. Mechanical Systems and Signal Processing, 2023, 199:1-15.
［10］尹晨, 周世超, 何建樑, 等.基于多源同步信号与深度学习的刀具磨损在线识别方法［J］. 中国机械工程, 2021, 32(20):2482-2491.
YIN Chen, ZHOU Shichao, HE Jianliang, et al. Tool Wear Online Recognition Method Based on Multi-source Synchronous Signals and Deep Learning［J］. China Mechanical Engineering, 2021, 32(20):2482-2491.
［11］PAN W J, QU H Y, SUN Y H, et al. A Deep Convolutional Neural Network Model with Two-stream Feature Fusion and Cross-load Adaptive Characteristics for Fault Diagnosis［J］. Measurement Science and Technology, 2023, 34(9):1-23.
［12］JADERBERG M, SIMONYAN K, ZISSERMAN A, et al. Spatial Transformer Networks［M/OL］. Cambridge:MIT Press,［2023-10-12］.https:∥doi.org/10.48550/arXiv.1506.02025.
［13］WANG W G, SHEN J B, DONG X P, et al. Salient Object Detection Driven by Fixation Prediction［C］∥IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake, 2018:1711-1720.
［14］HU J, SHEN L, ALBANIE S, et al. Squeeze and Excitation Networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8):2011-2023.
［15］谢松峰. CFRP/TC4叠层构件低频振动制孔刀具磨损状态监测技术研究［D］. 南京:南京航空航天大学, 2021.
XIE Songfeng. Research on the Monitoring Technology of Low-frequency Vibration Tool Wear State of CFRP/TC4 Laminated Components for Hole Making［D］. Nanjing:Nanjing University of Aeronautics and Astronautics, 2021.
［16］杨蕊, 李宏坤, 贺长波, 等. 利用最优小波尺度循环谱的滚动轴承早期故障特征提取［J］. 机械工程学报, 2018, 54(17):208-217.
TANG Rui, LI Hongkun, HE Changbo, et al. Early Failure Feature Extractionof Rolling Bearings Using Optimal Wavelet Scale Cyclic Spectr［J］. Journal of Mechanical Engineering, 2018, 54(17):208-217.
［17］许祖德. 国际标准ISO3685—1977(E)单刃车削刀具的寿命试验(续)［J］. 工具技术, 1979 (5):51-64.
XU Zude. International Standard ISO 3685-1977(E) Life Tests For Single Flute Turning Tools (continued)［J］. Tool Technology, 1979(5):51-64.
［18］LI G, HU J Y, SHAN D W, et al. A CNN Model Based on Innovative Expansion Operation Improving the Fault Diagnosis Accuracy of Drilling Pump Fluid End［J］. Mechanical Systems and Signal Processing, 2023, 187:1-15.
［19］李正官, 韩天杰, 王超群, 等. 基于VGG-19卷积神经网络的刀具磨损监测方法［J］. 机械设计与制造工程, 2020, 49(6):93-97.
LI Zhengguan, HAN Tianjie, WANG Chaoqun, et al. Tool Wear Monitoring Method Based on VGG-19 Convolutional Neural Network［J］. Mechanical Design and Manufacturing Engineering, 2020, 49(6):93-97.
［20］HE K M, ZHANG X Y, REN S Q, et al. Deep Residual Learning for Image Recognition［C］∥IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, 2016:770-778.

[1]	王秋莲1, 欧桂雄1, 徐雪娇1, 刘锦荣1, 马国红2, 邓红标2. 基于VMD-SSA-LSTM考虑刀具磨损的数控铣床切削功率预测模型研究[J]. 中国机械工程, 2024, 35(06): 1052-1063.
[2]	勾睿杰, 张晓峰, 张鸿滨, 姚俊, 李勋. 刀具磨损对Allvac 718Plus高温合金铣削加工表面完整性及疲劳性能的影响[J]. 中国机械工程, 2023, 34(24): 2920-2926.
[3]	李小睿, 赵威, 李浩, 史卫奇, 何宁. 高压低温CO2射流冷却条件下高速车削淬硬轴承钢的试验研究[J]. 中国机械工程, 2023, 34(24): 2975-2985.
[4]	刘会永, 张松, 李剑峰, 栾晓娜, . 采用改进CNN-BiLSTM模型的刀具磨损状态监测[J]. 中国机械工程, 2022, 33(16): 1940-1947，1956.
[5]	史珂铭, 邹益胜, 刘永志, 丁昆, 丁国富. 一种不同工艺条件下刀具磨损状态多类域适应迁移辨识方法[J]. 中国机械工程, 2022, 33(15): 1841-1849.
[6]	吴世雄, 张文锋, 刘广东, 王成勇. 低温液氮冷却下高速切削淬硬钢的切屑形成及刀具磨损[J]. 中国机械工程, 2022, 33(05): 551-559.
[7]	赵国龙, 信连甲, 李亮, 王珉, 郝秀清, 何宁. 高硅铝合金的金刚石涂层刀具铣削损伤机理研究[J]. 中国机械工程, 2022, 33(02): 153-159.
[8]	李聪波, 孙鑫, 侯晓博, 赵希坤, 吴少卿. 数字孪生驱动的数控铣削刀具磨损在线监测方法[J]. 中国机械工程, 2022, 33(01): 78-87.
[9]	尹晨, 周世超, 何建樑, 孙宇昕, 王禹林. 基于多源同步信号与深度学习的刀具磨损在线识别方法[J]. 中国机械工程, 2021, 32(20): 2482-2491.
[10]	戴稳1;张超勇1;孟磊磊1;薛燕社1;肖鹏飞1;尹勇2. 采用深度学习的铣刀磨损状态预测模型[J]. 中国机械工程, 2020, 31(17): 2071-2078.
[11]	何彦1;凌俊杰1;王禹林2;李育锋1;吴鹏程1;肖圳1. 基于长短时记忆卷积神经网络的刀具磨损在线监测模型[J]. 中国机械工程, 2020, 31(16): 1959-1967.
[12]	李友生. [刀具及切削工艺]高速铣削中TaC（NbC）对硬质合金刀具磨损性能的影响[J]. 中国机械工程, 2020, 31(13): 1601-1605.
[13]	柳培蕾;杨世强;白乐乐;李卓;李德信. 基于磨料磨损的盘形滚刀刀圈磨损模型[J]. 中国机械工程, 2019, 30(15): 1782-1789.
[14]	卢志远, 马鹏飞, 肖江林, 王美清, 唐晓青. [设备健康管理]基于机床信息的加工过程刀具磨损状态在线监测[J]. 中国机械工程, 2019, 30(02): 220-225.
[15]	王丽华1;杨家巍1;张永宏1;赵晓平2;谢阳阳1. 基于堆叠降噪自编码的刀具磨损状态识别[J]. 中国机械工程, 2018, 29(17): 2038-2045.